An M2M communication network consists of various M2M nodes and bearer networks. The M2M nodes implement communication through the bearer networks, and each M2M node at least includes an application or a CSE.
The M2M nodes may be divided into three categories according to different functions: application nodes, intermediate nodes and basic nodes, as shown in FIG. 1, wherein
each application node is required to include at least one M2M application, and does not include any CSE;
each intermediate node is required to include at least one CSE, and may include at least one M2M application; and
each basic node is required to include at least one CSE and at least one M2M application.
Communication among the M2M applications is implemented by interaction among the CSEs, the M2M applications are connected to the CSEs through X interfaces, and the CSEs communicate through Y interfaces, as shown in FIG. 1. In a practical application process, a local CSE sends a request message sent by an M2M application to a target CSE through the bearer network, but the request message may be invalidated immediately, thereby causing a sending failure and having influence on implementation of the M2M application when the bearer network fails or is disconnected. Moreover, after the bearer network is recovered, the M2M application is required to resend the request message to the local CSE, so that network signalling is correspondingly increased, and a network resource is consumed.